Image_harmony

calculate image color harmony by color hue degree

Library

• numpy
• opencv
• matplotlib.pyplot
• collections Counter
• scipy stats
• scipy.optimize find_peaks

How to use

You can use module like this

• image_harmony(image_path, threshold(default = 0.005), hue_distance(default = 15) ==> minimum harmony, harmony_graph(default = false))
• ex) image_harmony('./Lenna.png', threshold=0.005, hue_distance=15, harmony_graph=False)

Result

This function will return (Image_harmony, 1st peak xy, 2nd peak xy, peaks x, peaks y, Completion_time)

• Image_harmony : The higher value means more harmonious image
• ex) image_harmony('./Lenna.png', threshold=0.005, hue_distance=15, harmony_graph=False)

  • (28, [10, 0.01728678506424689], [342, 0.015386930437978728], [10, 342], [0.01728678506424689, 0.015386930437978728], 0.7515513896942139)

if it has no harmony value

• image_harmony('./Lenna.png', threshold=0.005, hue_distance=30, harmony_graph=False)

  • ('Color Inharmonic', [10, 0.01728678506424689], ['None', 'None'], [10, 342], [0.01728678506424689, 0.015386930437978728], 0.7615368366241455)

return graphs

1. color harmony graph